New Directions in Infection-Associated Ischemic Stroke.

The relationship between infections and stroke has not been fully characterized, probably delaying the development of specific treatments. This narrative review addresses mechanisms of stroke linked to infections, including hypercoagulability, endothelial dysfunction, vasculitis, and impaired thrombolysis. SARS-CoV-2, the virus that causes COVID-19, may promote the development of stroke, which may represent its most severe neurological complication. The development of specific therapies for infection-associated stroke remains a profound challenge. Perhaps the most important remaining issue is the distinction between infections that trigger a stroke versus infections that are truly incidental. This distinction likely requires the establishment of appropriate biomarkers, candidates of which are elevated levels of fibrin D-dimer and anticardiolipin/antiphospholipid antibodies. These candidate biomarkers might have potential use in identifying pathogenic infections preceding stroke, which is a precursor to establishing specific therapies for this syndrome.

A novel and simple approach to regularise attention frameworks and its efficacy in segmentation.

Deep neural networks with attention mechanism have shown promising results in many computer vision and medical image processing applications. Attention mechanisms help to capture long range interactions. Recently, more sophisticated attention mechanisms like criss-cross attention have been proposed for efficient computation of attention blocks. In this paper, we introduce a simple and low-overhead approach of adding noise to the attention block which we discover to be very effective when using an attention mechanism. Our proposed methodology of introducing regularisation in the attention block by adding noise makes the network more robust and resilient, especially in scenarios where there is limited training data. We incorporate this regularisation mechanism in the criss-cross attention block. This criss-cross attention block enhanced with regularisation is integrated in the bottleneck layer of a U-Net for the task of medical image segmentation. We evaluate our proposed framework on a challenging subset of the NIH dataset for segmenting lung lobes. Our proposed methodology results in improving dice-scores by 2.5 % in this context of medical image segmentation.

Attention-augmented U-Net (AA-U-Net) for semantic segmentation.

Deep learning-based image segmentation models rely strongly on capturing sufficient spatial context without requiring complex models that are hard to train with limited labeled data. For COVID-19 infection segmentation on CT images, training data are currently scarce. Attention models, in particular the most recent self-attention methods, have shown to help gather contextual information within deep networks and benefit semantic segmentation tasks. The recent attention-augmented convolution model aims to capture long range interactions by concatenating self-attention and convolution feature maps. This work proposes a novel attention-augmented convolution U-Net (AA-U-Net) that enables a more accurate spatial aggregation of contextual information by integrating attention-augmented convolution in the bottleneck of an encoder-decoder segmentation architecture. A deep segmentation network (U-Net) with this attention mechanism significantly improves the performance of semantic segmentation tasks on challenging COVID-19 lesion segmentation. The validation experiments show that the performance gain of the attention-augmented U-Net comes from their ability to capture dynamic and precise (wider) attention context. The AA-U-Net achieves Dice scores of 72.3% and 61.4% for ground-glass opacity and consolidation lesions for COVID-19 segmentation and improves the accuracy by 4.2% points against a baseline U-Net and 3.09% points compared to a baseline U-Net with matched parameters. Supplementary Information: The online version contains supplementary material available at 10.1007/s11760-022-02302-3.

Novel Insights on Induced Sparsity in Multi-Time Attention Networks.

Current deep learning approaches for dealing with sparse irregularly sampled time-series data do not exploit the extent of sparsity of the input data. Our work is inspired by the sparse and irregularly sampled nature of physiological time series data in electronic health records. We explore the effect of inducing varying degrees of sparsity on the predictive performance of Multi-Time Attention Networks (mTAN) [1]. Our methodology is to induce sparsity by first sub-sampling the time-series before feeding it to the mTAN network. We conduct empirical experiments with sub-sampling ranging from 10 to 90 %. We investigate the performance of our methodology on the Human Activity dataset and Physionet 2012 mortality prediction task. Our results demonstrate that our proposed time-point sub-sampling coupled with mTAN improves the performance by 2 % on the Human Activity dataset with 80 % lesser time-points for training. On the Physionet dataset, our approach achieves comparable performance as baseline with 30 % lesser time-points. Our experiments reveal that time-series data could be further coarsely acquired when used in tandem with state-of-the-art networks capable of handling sparse data (mTAN). This could be of immense help for various applications where data acquisition and labeling is a significant challenge.

Novel no-reference multi-dimensional perceptual similarity metric.

Enormous progress has been made in the domain of determining image quality. However, even the recently proposed deep learning based perceptual quality metrics and the classical structural similarity metric (SSIM) are not designed to operate in the absence of a good quality reference image. Many of the image acquisition processes, especially in medical imaging, would immensely benefit from a metric that can indicate if the quality of an image is improving or worsening based on adaptation of the acquisition parameters. In this work, we propose a novel multi-dimensional no-reference perceptual similarity metric that can compute the quality of a given image without a reference pristine quality image by combining no-reference image quality metric (PIQUE) and perceptual similarity. The dimensions of quality currently explored are in the axis of noise, blur, and contrast. Our experiments demonstrate that our proposed novel no-reference perceptual similarity metric correlates very well with the quality of an image in a multi-dimensional sense.

Deformable attention (DANet) for semantic image segmentation.

Deep learning based medical image segmentation is currently a widely researched topic. Attention mechanism used with deep networks significantly benefit semantic segmen-tation tasks. The recent criss-cross-attention module captures global self-attention while remaining memory and time efficient. However, capturing attention from only the pertinent non-local locations can cardinally boost the accuracy of semantic segmentation networks. We propose a new Deformable Attention Network (DANet) that enables a more accurate contextual information computation in a similarly efficient way. Our novel technique is based on learning the deformation of the query, key and value attention feature maps in a continuous way. A deep segmentation network with this attention mechanism is able to capture attention from germane non-local locations. This boosts the segmentation performance of COVID-19 lesion segmentation compared to criss-cross attention within aU-Net. Our validation experiments show that the performance gain of the recursively applied deformable attention blocks comes from their ability to capture dynamic and precise (wider) attention context. DANet achieves Dice scores of 60.17% for COVID-19 lesions segmentation and improves the accuracy by 4.4% points compared to a baseline U-Net.

Evaluation Tool to Diagnose Faults and Discrepancy in Semi-Automated or Manual Annotations in Ultrasound Cine Loops (Videos).

Good quality (annotated) data is one of the most important aspects of supervised deep learning. Tasks such as semantic segmentation have a huge data requirement in exchange for only satisfactory performance. Large-scale annotations spread across multiple annotators tends to create inconsistencies, as there are various manual and semi-automated techniques involved. This mandates an external evaluator or expert to check and narrow down the problematic annotations. Studies have shown that even marking a few instances wrong in classification can lead to a significant performance drop in the model (mislabeling only 10% of one class can degrade the total performance of all classes by up to 10%). It has been noticed that fault localization by a medical expert is one of the most expensive and time-consuming processes. In this paper, we propose a novel framework for detecting the inconsistencies in the annotation of every object/anatomy in a specific image. We leverage the power of semi-supervised deep learning models (STCN) to help produce high-quality data for AI segmentation algorithms. Evaluation using this algorithm has been shown to reduce annotation review time by at least 5 hours for just 1000 images, and the quality of ground truth data improved thereby increasing the performance of the model by almost 3%.

Descriptive analysis of Acute Ischemic stroke in COVID-19 patients through the course of the COVID-19 pandemic.

Coronavirus disease 2019 (COVID-19) has been associated with Acute Ischemic Stroke (AIS). Here, we characterize our institutional experience with management of COVID-19 and AIS. Baseline demographics, clinical, imaging, and outcomes data were determined in patients with COVID-19 and AIS presenting within March 2020 to October 2020, and November 2020 to August 2021, based on institutional COVID-19 hospitalization volume. Of 2512 COVID-19 patients, 35 (1.39%, mean age 63.3 years, 54% women) had AIS. AIS recognition was frequently delayed after COVID-19 symptoms (median 19.5 days). Four patients (11%) were on therapeutic anticoagulation at AIS recognition. AIS mechanism was undetermined or due to multiple etiologies in most cases (n = 20, 57%). Three patients underwent IV TPA, and three underwent mechanical thrombectomy, of which two suffered re-occlusion. Three patients had incomplete mRNA vaccination course. Fourteen (40%) died, with 26 (74%) having poor outcomes. Critical COVID-19 severity was associated with worsened mortality (p = 0.02). More patients (12/16; 75%) had either worsened or similar 3-month functional outcomes, than those with improvement, indicating the devastating impact of co-existing AIS and COVID-19. Comparative analysis showed that patients in the later cohort had earlier AIS presentation, fewer stroke risk factors, more comprehensive workup, more defined stroke mechanisms, less instance of critical COVID-19 severity, more utilization of IV TPA, and a trend towards worse outcomes for the sub-group of mild-to-moderate COVID-19 severity. AIS incidence, NIHSS, and overall outcomes were similar. Further studies should investigate outcomes beyond 3 months and their predictive factors, impact of completed vaccination course, and access to neurologic care.

Dynamic deformable attention network (DDANet) for COVID-19 lesions semantic segmentation.

Deep learning based medical image segmentation is an important step within diagnosis, which relies strongly on capturing sufficient spatial context without requiring too complex models that are hard to train with limited labelled data. Training data is in particular scarce for segmenting infection regions of CT images of COVID-19 patients. Attention models help gather contextual information within deep networks and benefit semantic segmentation tasks. The recent criss-cross-attention module aims to approximate global self-attention while remaining memory and time efficient by separating horizontal and vertical self-similarity computations. However, capturing attention from all non-local locations can adversely impact the accuracy of semantic segmentation networks. We propose a new Dynamic Deformable Attention Network (DDANet) that enables a more accurate contextual information computation in a similarly efficient way. Our novel technique is based on a deformable criss-cross attention block that learns both attention coefficients and attention offsets in a continuous way. A deep U-Net (Schlemper et al., 2019) segmentation network that employs this attention mechanism is able to capture attention from pertinent non-local locations and also improves the performance on semantic segmentation tasks compared to criss-cross attention within a U-Net on a challenging COVID-19 lesion segmentation task. Our validation experiments show that the performance gain of the recursively applied dynamic deformable attention blocks comes from their ability to capture dynamic and precise attention context. Our DDANet achieves Dice scores of 73.4% and 61.3% for Ground-glass opacity and consolidation lesions for COVID-19 segmentation and improves the accuracy by 4.9% points compared to a baseline U-Net and 24.4% points compared to current state of art methods (Fan et al., 2020).

The Cerebro-Renal System- Anatomical and Physiological Considerations.

The brain and kidney both uniquely are highly susceptible to vascular injury from shared vascular risk factors. However these are not sufficient to explain the complete extent of cerebrovascular disease especially small vessel disease in its myriad presentations that patients with chronic kidney disease manifest. They both require a large amount of blood supply to function optimally. Shared anatomical and physiological factors such as the presence of strain vessels, the local vascular autoregulation that control blood supply possible, results in the vulnerability of these organs to the vascular risk factors. Because it is a bidirectional system where each affects the other, it is best considered as a cerebro-renal unit.

Asymmetries in force matching are related to side of stroke in right-handed individuals.

Asymmetries in grasp force matching extend beyond quantifying a single measure of maximum grip strength and advance our application of side-specific treatment interventions. A cross sectional study design investigated grasp-force matching performance in right-handed individuals with a stroke and age-matched healthy controls. A visual representation of the 20% Maximum Voluntary Contraction (MVC) was matched in three conditions in the absence of visual feedback with the same (Ipsilateral Remembered - IR) or opposite hand (Concurrent - CC and Contralateral Remembered - CR). Greater overall relative error (RE) was found in contralateral compared to ipsilateral matching tasks. In the CR condition, post hoc analysis revealed significant differences between control and right hemisphere damage (RHD) group (95% CI [16.41-88.59]; p < 0.01) as well as left hemisphere damage (LHD) group and RHD (95% CI [23.4-95.09]; p < 0.01). Right hand matching relative error was 2.49 times larger in the RHD compared to the LHD group. Within the RHD group, matching errors were greater for the right than left hand in both contralateral conditions (95% CI [34.25-101.07]; p < 0.001). Individuals with RHD showed greater asymmetries in contralateral matching tasks compared to LHD and controls. More specifically, the RHD group had the greatest difficulty matching tasks with their right (non-paretic) than left (paretic) hand. In order to elucidate this asymmetry in the clinic the use of complementary grasp measures may be considered.

Apixaban for the treatment of cerebral venous thrombosis: A case series.

BACKGROUND: Venous thrombosis affecting cerebral veins and sinuses (CVT) is an uncommon neurological condition. Traditionally patients are treated with intravenous heparin followed by an oral vitamin K antagonist like warfarin. Direct oral anticoagulants (DOACs) may offer advantages over warfarin. There is evidence to demonstrate the effectiveness of both dabigatran and rivaroxaban. No data, however, has been published describing the use of apixaban in patients with CVT. METHODS: Report of three cases of CVT and review literature on available treatment options; efficacy and safety of novel oral anticoagulants in patients with systemic thrombosis. RESULTS: All patients presented with typical features of CVT. After confirming the diagnosis, they were acutely treated with heparin and later discharged on apixaban. During follow up visits, they tolerated apixaban well and did not have any bleeding complications. Follow up scans showed resolution of the thrombus and recanalization. CONCLUSION: CVT is an uncommon neurological condition and is often complicated by associated intraparenchymal hemorrhage. Although not recommended in current guidelines, apixaban may be a safe and effective option for the treatment of CVT.

Young Ischemic Stroke in Association with Ovarian Hyperstimulation Syndrome.

BACKGROUND: Ovarian hyperstimulation syndrome (OHSS) is a serious and potentially fatal complication of medical ovarian stimulation. Uncommonly, it is associated with thromboembolic complications with venous thrombosis being more common than arterial thromboembolic events. We present a case of cerebral infarction in the setting of severe OHSS secondary to in vitro fertilization treatment with no residual neurological deficits. MATERIALS AND METHODS: We also performed a review of previously published ischemic cerebral infarction and cerebral venous sinus thrombosis (CVST) cases associated with OHSS to evaluate common patterns in presentations, commonly affected central nervous system sites, trends for therapeutic options in these cases, and outcomes. CONCLUSION: We have included 27 cases of ischemic cerebral infarction and 7 cases of CVST previously published in English literature. We have included cases of central retinal artery occlusion in the ischemic cerebral infarction group, and central retinal vein occlusion in the CVST group. Mean ages of presentation were 31 ± 4.84 and 34 ± 4.90 years for ischemic cerebral infarction and CVST, respectively. Ischemic strokes commonly affect large cortical areas with unilateral weakness, aphasia, unilateral sensory changes, and visual field deficits being the common presentations. Middle cerebral artery (n = 7) is the common site of vascular occlusion where vascular imaging has been reported, followed by internal carotid artery occlusion (n = 5). OHSS cannot be considered a direct risk for pathogenesis, but OHSS is frequently associated with hyperviscosity, which may add to the risk factors.

Acute Onset Dystonia after Infarction of Premotor and Supplementary Motor Cortex.

OBJECTIVE: Poststroke dystonia is the second most common movement disorder after chorea and often has a delayed manifestation. Lesions of the contralateral lenticular nucleus, particularly the putamen, have been implicated in the pathogenesis of dystonia. We present an unusual case of rapid onset of focal dystonia of the left upper extremity, which developed after infarction of the right premotor cortex (PMC) and the supplementary motor area (SMA). METHOD: A retrospective chart review of the patient was performed. RESULTS AND CONCLUSION: We propose that disruption of the afferents from PMC and SMA in the setting of chronic striatal abnormality can result in acute dystonia due to disinhibition of the thalamocortical circuit.